Expose audio analysis methods via API

shauneccles · Feb 21, 2024 · c6219cd · c6219cd
1 parent 3ffb12e
commit c6219cd
Show file tree

Hide file tree

Showing 4 changed files with 70 additions and 12 deletions.
diff --git a/ledfx/api/config.py b/ledfx/api/config.py
@@ -183,7 +183,6 @@ async def put(self, request: web.Request) -> web.Response:
         """
         try:
             config = await request.json()
-
             audio_config = validate_and_trim_config(
                 config.pop("audio", {}),
                 AudioInputSource.AUDIO_CONFIG_SCHEMA.fget(),
@@ -200,7 +199,6 @@ async def put(self, request: web.Request) -> web.Response:
             core_config = validate_and_trim_config(
                 config, CORE_CONFIG_SCHEMA, "core"
             )
-
             self._ledfx.config["audio"].update(audio_config)
             self._ledfx.config["melbanks"].update(melbanks_config)
             self._ledfx.config.update(core_config)

diff --git a/ledfx/api/schema.py b/ledfx/api/schema.py
@@ -6,7 +6,7 @@
 from ledfx.api import RestEndpoint
 from ledfx.api.utils import PERMITTED_KEYS, convertToJsonSchema
 from ledfx.config import CORE_CONFIG_SCHEMA, WLED_CONFIG_SCHEMA
-from ledfx.effects.audio import AudioInputSource
+from ledfx.effects.audio import AudioAnalysisSource, AudioInputSource
 from ledfx.effects.melbank import Melbank, Melbanks
 
 _LOGGER = logging.getLogger(__name__)
@@ -119,15 +119,32 @@ async def get(self, request: web.Request) -> web.Response:
 
             elif schema == "audio":
                 # Get audio schema
+                audio_input_schema = convertToJsonSchema(
+                    AudioInputSource.AUDIO_CONFIG_SCHEMA.fget()
+                )
+                audio_analysis_schema = convertToJsonSchema(
+                    AudioAnalysisSource.CONFIG_SCHEMA
+                )
+
+                merged_schema = {**audio_input_schema, **audio_analysis_schema}
+
+                for key in (
+                    audio_input_schema.keys() & audio_analysis_schema.keys()
+                ):
+                    if isinstance(
+                        audio_input_schema[key], dict
+                    ) and isinstance(audio_analysis_schema[key], dict):
+                        merged_schema[key] = {
+                            **audio_input_schema[key],
+                            **audio_analysis_schema[key],
+                        }
+
                 response["audio"] = {
                     "schema": {
-                        **convertToJsonSchema(
-                            AudioInputSource.AUDIO_CONFIG_SCHEMA.fget(),
-                        ),
+                        **merged_schema,
                         **{"permitted_keys": PERMITTED_KEYS["audio"]},
                     }
                 }
-
             elif schema == "melbanks":
                 # Get melbanks schema
                 response["melbanks"] = {

diff --git a/ledfx/api/utils.py b/ledfx/api/utils.py
@@ -18,7 +18,14 @@
 }
 
 PERMITTED_KEYS = {
-    "audio": ("min_volume", "audio_device", "delay_ms"),
+    "audio": (
+        "min_volume",
+        "audio_device",
+        "delay_ms",
+        "pitch_method",
+        "onset_method",
+        "pitch_tolerance",
+    ),
     "melbanks": (
         "max_frequencies",
         "min_frequency",

diff --git a/ledfx/effects/audio.py b/ledfx/effects/audio.py
@@ -110,7 +110,8 @@ def AUDIO_CONFIG_SCHEMA():
         default_device_index = AudioInputSource.default_device_index()
         valid_device_indexes = AudioInputSource.valid_device_indexes()
         input_devices = AudioInputSource.input_devices()
-        # melbanks = Melbanks.CONFIG_SCHEMA
+        melbanks = Melbanks.CONFIG_SCHEMA
+        audio_analysis = AudioAnalysisSource.CONFIG_SCHEMA
         return vol.Schema(
             {
                 vol.Optional("sample_rate", default=60): int,
@@ -165,6 +166,7 @@ def update_config(self, config):
                     self.input_devices()[self._config["audio_device"]]
                 )
             )
+        self._ledfx.config["audio"] = self._config
 
     def activate(self):
         if self._audio is None:
@@ -440,12 +442,46 @@ def volume(self, filtered=True):
 
 
 class AudioAnalysisSource(AudioInputSource):
+    # https://aubio.org/doc/latest/pitch_8h.html
+    PITCH_METHODS = [
+        "yinfft",
+        "yin",
+        "yinfast",
+        "fcomb",
+        "mcomb",
+        "schmitt",
+        "specacf",
+    ]
+    # https://aubio.org/doc/latest/specdesc_8h.html
+    ONSET_METHODS = [
+        "energy",
+        "hfc",
+        "complex",
+        "phase",
+        "wphase",
+        "specdiff",
+        "kl",
+        "mkl",
+        "specflux",
+    ]
     CONFIG_SCHEMA = vol.Schema(
         {
-            vol.Optional("pitch_method", default="default"): str,
+            vol.Optional(
+                "pitch_method",
+                default="yinfft",
+                description="Method to detect pitch",
+            ): vol.In(PITCH_METHODS),
             vol.Optional("tempo_method", default="default"): str,
-            vol.Optional("onset_method", default="specflux"): str,
-            vol.Optional("pitch_tolerance", default=0.8): float,
+            vol.Optional(
+                "onset_method",
+                default="hfc",
+                description="Method used to detect onsets",
+            ): vol.In(ONSET_METHODS),
+            vol.Optional(
+                "pitch_tolerance",
+                default=0.8,
+                description="Pitch detection tolerance",
+            ): vol.All(vol.Coerce(float), vol.Range(min=0.0, max=2)),
         },
         extra=vol.ALLOW_EXTRA,
     )